Vol. 168, No. 2, 1990 April 30, 1990

ENHANCED

BIOCHEMICAL

BINDING

ACTIVITY

AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 409-414

OF AN APOLIPOPROTEIN

APO E5, TO LDL RECEPTORS Li-Ming

ON HUMAN

Dong, Taku Yamamura’

E MUTANT,

FIBROBLASTS

and Akira Yamamoto

Department of Etiology and Pathophysiology, National Cardiovascular Center Research Institute, 5-7-l Fujishiro-dai, Suita, Osaka, 565 Japan Received

March

5, 1990

SUMMARY. We had previously demonstrated an apolipoprotein E (apo E) mutant, apo E5, associated with hyperlipidemia and atherosclerotic vascular diseases. To investigate the possible mechanism of apo E5 involvement in the development of hyperlipidemia, we have isolated the apo E isoprotein and determined its binding activity to LDL (low density lipoprotein) receptors. It was shown that the apo E5 isoprotein was two times more active for binding to LDL receptors than apo E3. The concentration of apo E, at which 50% of luIlabeled LDL bound to the receptor was displaced, was 29 rig/ml for apo E5 and 63 rig/ml for apo E3. This result suggests that the high affinity of apo E5 to LDL receptors results in a high uptake of apo ES-containing lipoproteins by the liver and leads to a down-regulation of LDL receptors in the liver. We can postulate that individuals with apo E5 are susceptible to hypercholesterolemia and in consequence to atherosclerosis. OlY30Academic Press, Inc.

Apolipoprotein

E (apo E) is one of the major apolipoprotein

components in very low

density lipoproteins (VLDL) and plays an important role in lipoprotein metabolism.

Apo E

has three major isoproteins, E4, E3 and E2, and several accompanying components, E4s, E3s and E2s, arising from posttranslational modification with sialic acids [1,21. Apo E4 and apo E2 result from a parent type of apo E3 due to the interchange of arginine and cysteine at residue 112 and 158, respectively [3,4]. Because of this interchange, they have isoelectric. points more basic and more acidic than apo E3 by one unit of charge, respectively. Receptor binding studies showed that apo E can bind to low density lipoprotein (LDL) receptors (apo B,E receptor) like apo B and can also bind to apo E specific hepatic lipoprotein receptors (apo E receptor) [5]. Apo E2, however, has a defective binding ability to lipoprotein

receptors [6,7], which results in the accumulation of remnants and decreased

LDL concentration in plasma due to the disturbance of lipoprotein metabolism through the ‘To whom correspondence should be addressed. Abbreviations used are: Apo E, Apolipoprotein E; VLDL, Very low density lipoprotein; LDL, low density lipoprotein; DMPC, Dimyristoylphosphatidylcholine; D’lT, Dithiothreitol.

409

0006-291X/90 $1.50 Copyright 0 1990 by Academic Press. Inc. All rights of reproduction in any form reserved.

Vol. 168, No. 2, 1990

BIOCHEMICAL

lipoprotein receptors in the liver [8,9].

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

In some cases, apo E2 homozygosity leads to type

III hyperlipoproteinemia 191. On the other hand, apo E4 was suggested to have an action opposite to that of apo E2 in lipoprotein metabolism. Individuals with apo E4 have hypercholesterolemia resulting from increased LDL levels in plasma compared to those with apo E3 [lO,ll]. We have previously demonstrated two unique isoproteins of apo E (apo E5 and E7) associated with hyperlipidemia and atherosclerosis [12,13]. According to isoelectric focusing, the apo E5 has two more positive charges than apo E3. Gene analysis demonstrated that it had an amino acid substitution (Glu,+Lys)

at the third residue from the amino terminus [14].

In this study, we characterized the activity of apo E5 binding to LDL receptors on human fibroblasts to investigate hyperlipidemia

the metabolism

of apo E.5 and its association with

and atherosclerosis.

MATERIALS

AND METHODS

Sera were obtained from hyperlipidemic Lipoprotein and apolipoprotein preparation: subjects with an E2/2 phenotype and an E3/3 phenotype and from the O-kindred family members with an E5/3 phenotype reported in reference [12]. The VLDL fractions were separated by ultracentrifugation and then delipidated with acetone/ethanol (1: 1). After VLDL apolipoproteins were solubilized in 0.1 M Tris-HCl (pH 7.4) containing 6 M guanidine, 0.01% EDTA and 1% Zmercaptoethanol, apo E was isolated by Sephacryl S-300 (Pharmacia) column chromatography (1.6 cm x 200 cm). The apo E fractions were collected and dialyzed against 5 mM NH,HCO, and then lyophilized. Isolation of apo E isoprotein: Preparative isoelectric focusing was used for the isolation of asialo apo E2 and apo E3 isoproteins. The apo E preparations from column chromatography were dissolved in 10 mM Tris-HCl (pH 8.6) containing 8 M urea and 10 mM dithiothreitol (DTT) and then subjected to isoelectric focusing. Focusing was performed on a solubilizable polyacrylamide gel (15 cm x 14 cm x 0.3 cm) containing 8 M urea and 2% Ampholine pH 5-7 (LKB Produktor). To prepare the solubilizable polyacrylamide gel, the acrylamide was polymerized with the disulflde-containing crosslinking reagent N,N’-bis-acrylylcystamine (Bio-Rad) instead of N,N’-methylene-bis-acrylamide [151. After isoelectric focusing for 20 hours, the protein bands were visualized by soaking the gel in water [16], and the opaque band in the gels corresponding to asialo apo E2 or apo E3 was cut out. The gel slice was solubilized by adding the reducing agent DTT, and the solution was applied to a column of heparin-Sepharose CL-6B (Pharmacia). Apo E isoproteins were separated from the acrylamide monomers by heparin affinity chromatography [12] and then dialyzed and lyophilized. For isolating the asialo apo E5 isoprotein, apo VLDL from the patients with the E5/3 phenotype was directly subjected to preparative isoelectric focusing as described above because only limited materials could be obtained. Analytical isoelectric focusing was carried out as previously described [ 131. Preparation of apo E-dimyristoylphosphatidylcholine: Apo E.dimyristoylphosphatidylcholine (DMPC, Sigma) complexes were prepared ‘as described by Rall et al. [17]. Briefly, apo E and a sonicated preparation of DMPC were mixed and incubated for 1 hour at 25 “C. The complexes were isolated after ultracentrifugation on a KBr density gradient (d=1.006-1.21 g/ml) at 49,000 rpm (RPS 50-2 rotor, Hitachi-Koki) for 20 hours at 15 “C. The isolated complexes had an average phospholipid/protein ratio of 3.7:1 (wt/wt). 410

Vol.

166,

No.

2, 1990

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

Cells and assay: Human ‘%labeled LDL was prepared by the method of Fielding et al. [18]. Normal human fibroblasts were cultured and competitive binding of apo E.DMPC with lBI-labeled LDL was assayed at 4 “C on ice as described by Innerarity et al. [ 191.

RESULTS

AND DISCUSSION

We found apo ES in three unrelated patients with hyperlipidemia our lipid clinic and in the coronary care unit [12].

and atherosclerosis at

To investigate the role of apo ES in

lipoprotein metabolism and the development of atherosclerosis, the binding activity of apo E5 to lipoprotein specific receptors was determined. Asialo apo E isoproteins, apo E2, apo E3 and apo E5, were isolated from human apo VLDL by preparative isoekctric focusing on solubilizable polyacrylamide gels. Analytical isoelectric focusing patterns of isolated apo E isoproteins are shown in Figure 1. Each apo E isoprotein was recombined with DMPC vesicles, and the receptor binding activity was estimated by measuring the competition by apo EeDMPC against the binding of ‘?-labeled

LDL to LDL receptors on human fibroblasts.

Apo E2mDMPC was defective in its binding to LDL receptors, compared to the wild form, apo E3 (Figure 2), and this result is identical with the report by Weisgraber et al. [6]. Apo ES.DMPC

had a binding ability similar to that of the parent type of apo E3 at the

indicated concentrations (Figure 2).

On the other hand, when we tested their ability to compete with LDL at lower concentrations of apo E, we found that apo E5 showed a higher binding activity than apo E3 (Figure 3). A logit-log plot of the binding data was used to

1

2

3

4

5

6

F’ we 1. Analytical isoelectric focusing patterns on polyacrylamide gels (pH 3.5-8) of LDL (lanes 1, 3, and 5) and isolated asialo apo E isoproteins (lanes 2, 4, and 6). The apo E phenotypes shown are: (1) E2/2, (3) E3& and (5) ES/3. The isolated apo E isoproteins shown are: (2) asialo apo E2, (4) asialo apo E3, and (6) asialo apo ES. The apo ES isoprotein was isolated from O-kindred family members with an ES/3 phenotype, as reported in reference [12]. The four apoprotein bands in the lower portion of gels 1, 3 and 5 are apo C isoforms. The cathode (basic) is at the top and the anode (acidic), at the bottom.

*

411

BIOCHEMICAL

Vol. 168, No. 2, 1990

AND BIOPHYSICAL RESEARCH COMMUNICATIONS ,

100

5 f 6 c 1 4 f

0

2

Apo-PDMPC

”a

9.

0.5

1.0

Complexes

0.05 0

@g protein/ml)

3

Ape-E*DMPC

Complexes

0.10 +J protein/ml)

Figure 2. Ability of apo E*DMPC complexesof apo E2 ( A ), apo E3 ( 0 ) and ape E5 ( l ) to competewith T-labeled LDL for binding to normal human fibroblasts. After incubationfor 48 hoursin Dulbecco’smodifiedEagle’smediumcontaining10% humanlipoprotein-deficient serum,tbe cellsreceived2 ml of the samemediumwith 2 @ml of ‘?-labeled LDL aud the indicatedconcentrations of apoE.DMPC complexes.After two hoursincubationat 4 “C on ice, tbe cells wereextensivelywashedaud the ‘2sI-labe1ed LDL boundto the cells wasdetermined.The 100%control value was60 ng of ‘ZSI-labeled LDL proteinbound/mgcellularprotein. Eachpoint indicatestbe averageof duplicatedishes,Ape E2 ( A) wasdefectivein its binding activity, while both apoE3 (0) and apoES (0) bad similarbindingactivitiesat the indicatedconcentrations. Figure 3. Comparisonof the binding activities of apo E3eDMPCcomplexes( 0) and apo ESDMPC complexes( l ) to LDL receptorson human fibroblasts. The experimentswere cartied out as describedin Figure 2. The 100%control value was 50 ng of ‘2SI-Iabeled LDL protein bound/mgof cellular protein. The inset showsa logit-log plot of the binding data that was usedto determinetbe 50% displacementof lBI-labeledLDL. Tbe concentrationof apoE at which 50% of ISI-labeledLDL wasdisplaced was29 rig/ml for apo E5 ( l ) and 63 rig/ml for apo E3 ( 0).

determine the concentration required for 50% competition [6,20]. The concentration of apo EDMPC at which 50% of the ‘2SI-labeledLDL was displaced was 29 rig/ml for apo E5 and 63 rig/ml for apo E3. In a separateseriesof studies utilizing the same binding assay with different apo E.DMPC complexes, ‘2SI-labeledLDL preparationsand cell culture, we obtained a similar result: 30 rig/ml and 66 @ml, respectively. Apo E5 is two times more active in its binding to LDL receptors than apo E3. Recently, several clinical studies have shown that apo E2 seemsto be associatedwith hypocholesterolemia due to decreased LDL levels [g-11] and apo E4, associated with hypercholesterolemiadue to increased LDL levels [lo,1 11. In vifro investigations demonstrated apo E2 had markedly impaired binding activity to both LDL receptors and apo E receptors [6,7]. This will lead to a low conversion rate of VLDL remnantsto LDL, and to a decreasedtransport of chylomicron remnants to the liver, resulting in an up-regulation of hepatic LDL receptors. As a consequence,individuals with apo E2 may have a reduced A13

Vol. 168, No. 2, 1990

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

plasma LDL concentration. There are several novel apo E isoproteins in type III hyperlipoproteinemia, such as apo E2’(Arg,,+Cys) [ 171, apo E2”(Lys,,+Gln) [21], or apo E&&Arg,,,+Ser) E2(Arg,,,+Cys)

[4].

[22], and they have different primary structures from the common apo Although all of them showed a reduction in their binding to LDL

receptors, their binding activities differed from each other, corresponding to their structural differences [6,17,21,22].

In addition to the naturally occurring mutants of apo E exhibiting

defective receptor binding, a mutant apo E(Ser139+Arg, Leu,,+Ala)

created by site-specific

mutagenesis was reported to display slightly enhanced receptor binding activity [23]. From these observations, it has been suggested that the binding domain of apo E exists near residues 136-158 of the polypeptide chain and that the basic amino acids around the binding domain play an important role in the interaction with the lipoprotein receptors. Furthermore, the 171183 region of apo E was postulated to stabilize or align the receptor binding domain, and the importance of the conformation for receptor binding was suggested [24]. While apo E4 (with one additional positive charge derived from an amino acid substitution at a residue not located in its binding domain) shows sufficient binding activity to LDL receptors [6], it seems to have an effect opposite to that of apo E2, which results in an elevation of plasma LDL concentration [lO,ll]. In the present irr vitro experiments, the mutant apo E5 isoprotein showed two times higher activity for binding to LDL receptors on human fibroblasts than apo E3. Assuming that apo E5 has the same binding activity to LDL receptors and apo E receptors on hepatocytes, it could have the ability to transport chylomicron

remnants to the liver at a

higher rate, leading to a down-regulation of LDL receptors in the liver, and to convert VLDL This would remnants to LDL rapidly in receptor-mediated lipoprotein metabolism. consequently increase the concentration of plasma LDL and result in hypercholesterolemia. Indeed, many patients with apo E5 have had hypercholesterolemia [12] and could then be susceptible to atherosclerosis. Further studies on apo E5 are needed for a fuller understanding of its association with hyperlipidemia

and atherosclerosis.

ACKNOWLEDGMENTS

We wish to thank Drs. S. Tajima and Y. Miyake, National Cardiovascular Center Research Institute, for their many helpful suggestions. This work was supported in part by a Grant-in-Aid for Scientific Research (No. 61440053) from the Ministry of Education, Science and Culture of Japan, and grants from the Takeda Medical Research Foundation (1986-88) and the Yamanouchi Foundation for Research on Metabolic Disorders (1988). REFERENCES

1. Zannis, V.I., and Breslow, J.L. (1981) Biochemistry 20, 1033-1041. 2. Zannis, V.I., Breslow, J.L., Utermann, G., MahIey, R.W., Weisgraber, K.H., Havel, R.J., Goldstein, J.L., Brown, M.S., Schonfeld, G., Hazzard, W.R., and Blum, C. (1982) J. Lipid Res. 23, 911-914. 413

vol.

168, No. 2, 1990

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

3. 4. 5. 6.

Weiqraber, K.H., Rall, S.C.,Jr., and Mahley, R.W. (1981) J. Biol. Chem. 256, 9077-9083. Rall, S.C.,Jr., Weisgraber, K.H., and Mahley, R.W. (1982) J. Biol. Chem. 257, 4171-4178. Mahley, R.W., and Innerarity, T.L. (1983) Biochim. Biophys. Acta 737, 197-222. Weisgraber, K.H., Innerarity, T.L., and Mahley, R.W. (1982) J. Biol. Chem. 257, 2518-2521. 7. Hui, D.Y., Innerarity, T.L., and Mahley, R.W. (1984) J. Biol. Chem. 259, 860-869. 8. Utermann, G., Pruin, N., and Steinmetz, A. (1979) Clin. Genet. 15, 63-72. 9. Utermann, G., Hees, M., and Steinmetz, A. (1977) Nature 269, 604-607. 10. Ehnholm, C., Lukka, M., Kuusi, T., Nikkila, E., and Utermann, G. (1986) J. Lipid Res. 27, 227-235. 11. Sing, C.F., and Davignon, J. (1985) Am. J. Hum. Genet. 37, 268-285. 12. Yamamura, T., Yamamoto, A., Hiramori, K., and Nambu, S. (1984) Atherosclerosis 50, 159-172. 13. Yamamura, T., Yamamoto, A., Sumiyoshi, T., Hiramori, K., Nishioeda, Y., and Nambu, S. (1984) J. Clin. Invest. 74, 1229-1237. 14. Tajima, S., Yamamura, T., and Yamamoto, A. (1988) J. Biochem. 104, 48-52. 15. Hansen, J.N., Pheiffer, B.H., and Boehnert, J.A. (1980) Anal. Biochem. 105, 192-201. 16. von Hungen, K., Chin, R.C., and Baxter, CF. (1983) Anal. Biochem. 128, 398404. 17. Rail, S.C.,Jr., Weisgraber, K.H., Innerarity, T.L., and Mahley, R.W. (1982) Proc. Natl. Acad. Sci. USA 79, 4696-4700. 18. Fielding, C.J., Vlodavsky, I., Fielding, P.E., and Gospodarowicz, D. (1979) J. Biol. Chem. 254, 8861-8868. 19. Innerarity, T.L., Pitas, R.E., and Mahley, R.W. (1979) J. Biol. Chem. 254, 4186-4190. 20. Rodbard, D., Bridson, W., and Rayford, P.L. (1969) J. Lab. Clin. Med. 74, 770-781. 21. Rail, S.C.,Jr., Weisgraber, K.H., Innerarity, T.L., Bersot, T.P., and Mahley, R.W. (1983) J. Clin. Invest. 72, 1288-1297. 22. Wardell, M.R., Brennan, S.O., Janus, E.D., Fraser, R., and Carrell, R.W. (1987) J. Clin. Invest. 80, 483-490. 23. Lalazar, A., Weisgraber, K.H., Rall, S.C.,Jr., Giladi, H., Innertity, T.L., Lavanon, A.Z., Boyles, J.K., Amit, B., Gore&i, M., Mahley, R.W., and Vogel, T. (1988) J. Biol. Chem. 263, 3542-3545. 24. Lalazar, A., Ou, S.-H.I., and Mahley, R.W. (1989) J. Biol. Chem., 264, 8447-8450.

did